Cited, not ranked.

Here is how we currently describe ourselves across our properties: [paste your website boilerplate, LinkedIn description, Crunchbase entry, any directory listings].
Our legal name is [name], we operate as [trading name if different], and our category is [category].
Identify every inconsistency in name, category, and description across these sources. Flag any name collision with another company that a model would plausibly confuse us with.
Then produce the single canonical block — name, one-line category, two-sentence description — that should appear identically in all of them, and list every property I need to update to match.

Our category is [category] and our buyer is [buyer role, industry, company size].
Find the directories, category roundups, comparison sites, analyst listings, and review platforms that assistants are most likely to draw on when answering “best [category] for [buyer].”
For each: how an entry gets created, whether it is free or paid, whether it is editorially curated or self-serve, and how long inclusion typically takes.
Rank them by likely influence on a model's answer, not by ease of entry, and tell me which three to pursue first.

Run this query in three assistants: “Who are the leading vendors for [category] for [buyer type]?”
Here are the answers: [paste all three].
For each competitor named: what specific evidence is the model citing, and where does that evidence appear to come from?
Tell me which of those evidence types we have but have not published, which we could produce within a quarter, and which we cannot credibly claim at all.

Here is proprietary data we hold but have not published: [deployment counts, benchmark results, field or clinical outcomes, aggregate customer metrics, internal research].
Our buyer is [buyer role] and the decision they are making is [decision].
For each item, tell me: is it specific enough that a model would quote it verbatim, and does it bear on the decision the buyer is actually making?
For the ones that qualify, write the single sentence — with the number in it — that we should publish. Then flag anything I would need legal or customer approval to release.

Our credentials in our home market are: [clearances, certifications, marquee clients, deployment numbers, revenue].
We are selling into [target market] to [buyer role].
For each credential, tell me: does it carry weight with this buyer as stated, does it need restating in a local equivalent, or does it carry no signal at all? For the ones that need restating, write the sentence I should use instead.
Then list what proof this buyer would expect that we do not currently have.

Here is a page from our site: [paste the content].
Our buyer is [buyer role].
List the questions this page should be the cited answer to, phrased the way a buyer would type them into an assistant.
Then restructure the content so each question is a heading with its answer in the first two sentences beneath it, specifics before context. Convert any comparison or specification prose into a table.
Flag every claim on the page that has no number, source, or date attached to it.

Our site is [URL] and our category is [category]. Our key pages are [list].
Draft an llms.txt that tells an assistant what we do, who we are for, and which pages carry the authoritative answers.
Then specify the JSON-LD we should add: Organization, Product, and FAQPage schema, with the fields that matter for our category.
Finally, list what an evaluating agent would need from us that we do not currently expose — pricing, specification detail, integration documentation, an API — and rank those by how often they would block a shortlist decision.